The Lack of PPARα and Glucagon Receptors Immunizes the SCN Central Clock Against RF-Induced Metabolic Perturbations.

The SCN expression of Period 1 and 2 (Per1 and Per2) was reported to be unaltered after 9 d of RF (RF9) (4). Of note, transcript analyses on RF15 and RF30 failed to detect any variation in the expression of not only Per1 and Per2, but also of the other circadian clock components RORα, Bmal1, Cry1, and RevErbα (Fig. 2A and Fig. S2A). This raised the possibility that the SCN CC may either lack or be unresponsive (due to the absence of their transducers) to the RF metabolic signals triggering the shift of PCCs (5). Most notably, on quantitative RT-PCR (qRT-PCR) of microdissected SCN, we could not detect any transcripts for the peroxisome proliferator-activated receptor alpha (PPARα) and glucagon (subunit GαS) receptors, which orchestrate the CC shift in peripheral tissues (Fig. 2B) (5). Accordingly, in situ hybridization of SCN sections of both control and RF mice failed to reveal PPARα transcripts (Fig. 2C). In marked contrast, PPARα expression was readily detected in the hippocampus and paraventricular nucleus (PVN), in which a peripheral 12-h CC shift was also observed (Fig. 2B and Fig. S2 B and C). These CC shifts in PVN and hippocampus are noteworthy, as (i) the PVN, which is connected with the SCN via both afferent and efferent projections, has been proposed to function as a “relaying center” for SCN-generated signals (13), and (ii) the proper functioning of peripheral circadian clocks in the hippocampus is known to facilitate memory formation (14), which suggests that RF could interfere with memory processes (15).

Open in a separate window

Fig. 2.

The SCN CC is insensitive to the RF-induced metabolic alterations. (A) RNA transcript levels of CC components in the SCN of control and RF30 mice. (B) RNA transcript levels of PPARα and glucagon receptors in the SCN of control and RF30 mice. Expression of these genes in liver was evaluated as a control. (C) In situ hybridization of brain sections of control and RF mice with PPARα riboprobes. RORα expression was used as a marker for localizing the SCN. (D) Actimetric analyses of the circadian locomotor activity in control and RF30 mice. All values are mean ± SEM.

Open in a separate window

Fig. S2.

The central SCN CC is unaffected by the RF. (A) RNA transcript levels of CC components in the SCN of control and RF15 mice. (B) RNA transcript levels of CC components in the PVN of control and RF15 mice. (C) RNA transcript levels of CC components genes in the hippocampus (dentate gyrus) of control and RF30 mice. (D) RNA transcript level of GR in the hippocampus (dentate gyrus) and SCN of control and RF30 mice. All values are mean ± SEM.

Importantly and in keeping with the established role of the SCN in controlling the diurnal alternance of active (ZT12–ZT0) and rest (ZT0–ZT12) phases (1–3), analyses of the locomotor activity of RF mice revealed that, on RF, this alternance was unaffected (Fig. 2D). Thus, by excluding the expression of two transducers (PPARα and glucagon receptor) necessary to induce the PCC shift (5), the neurons of the SCN ensure the constancy of the master SCN CC during RF, as well as the proper control of the diurnal alternance of active and rest phases (1–3).

From the teleological perspective, this insensitivity of the SCN CC during a RF starvation-like state is an expected necessity, as the SCN-controlled active phase has to occur during the mouse dark cycle (the human light cycle), even though, due to the misalignment of the PCCs, this active period corresponds to the rest phase in peripheral tissues. This evolutionary design, which maintains the original active phase (wakefulness) in the SCN CC, as well as the rest phase (sleep) during RF-like conditions, is indeed advantageous as it provides the opportunity for a starving organism to efficiently look for food during the normal active phase, thereby increasing the chance to put an end to starvation and to readily realign the PCCs with the unchanged SCN master clock, as soon as feeding is restored.

It was previously suggested that the RF insensitivity of the SCN CC is related to the lack of the glucocorticoid (GC) receptor (GR) in SCN (16), as such a lack (that we confirmed; Fig. S2D) would prevent a SCN CC shift through an alteration of Per1/Per2 expression on RF extra-corticosterone production (16, 17). Our present results do not exclude this possibility. However, we found that the lacks of both PPARα and glucagon receptors in SCN are sufficient on their own to prevent a shift of the whole SCN CC on RF, as taken together they prevent the RF-induced CC shifts of RevErbα through PPARα increase and of Per1/Per2 through glucagon/CREB activation (5).

In RF Mice, the Expression of PCC-Controlled Output Genes Is Shifted by 12 h with Respect to the Diurnal Active and Rest Phases Controlled by the Central SCN CC.

Under physiological conditions, the level of expression of nearly 10–15% of the genes expressed in mice is controlled by PCCs, such that specific sets of genes are selectively expressed during the active and rest phases, their expression being directly controlled by CC components or indirectly by their output genes (1–3). Notably, it has been demonstrated that the normal expression of RORE DBS-bearing genes in different tissues is at their zenith during the circadian active phase (while at nadir during the rest phase), whereas E- and D-Box DBS-bearing genes are at their zenith during the circadian rest phase (18–20).

As the RF CC shifts occur in peripheral tissues, but not in SCN, we investigated whether the misalignment between the unchanged SCN CC-controlled genes expressed during the active and rest phases will lead to an “out of phase” expression of PCC-controlled output genes, i.e., whether the PCC active phase genes would be expressed during the SCN rest phase, whereas the PCC rest phase genes would be expressed during the SCN active phase. A bioinformatic search in the mouse genome for genes harboring D-Box DBS in their promoter-enhancer regions revealed numerous candidates, of which ~2,000 have human orthologs (Dataset S1 and SI Methods). Among these, we chose 40 D-Box–containing genes having known functions. We also chose 40 RORE DBS-containing genes all having known homeostatic functions (19). When analyzed at RF15 in liver and IECs the circadian expression of such RORE DBS- and D-Box–containing genes, which in WT mice also displayed a circadian variation (Tables S1–S4), confirmed a 12-h shift (misalignment) in the expression, such that the expression of the normally active phase-restricted RORE genes was shifted to the rest phase (Tables S1 and ​andS2),S2), whereas the initially rest phase-restricted D-Box genes were expressed during the active phase (Tables S3 and ​andS4).S4). Importantly, this misaligned expression of numerous CC-controlled output genes (Tables S1–S4), including those of endocrine factors (INS, IGF1, and FGF21), key transcription factors (c-Jun, IRF3, ATF5, and FXRβ), critical enzymes (NAMPT, Hsd3b5, FAS, and HMGCR), receptors, and transporters (INSR, IRS2, Glut2, and TLRs), all controlling essential physiological homeostatic processes, sets the stage for the development of RF-associated pathologies (see below).

Mice Selectively Fed During the Rest Phase Display a Metabolic Syndrome-Like Pathology due to the Misalignment of the Peripheral Clocks with the Central SCN CC.

The physiological alignment of PCCs with the master central SCN CC has emerged as an important factor for the homeostatic maintenance of an organism, as misalignments of PCCs with the SCN CC, such as those associated with shift work and RF, both of which correspond to activities performed during the SCN CC-controlled rest phase while being physiologically controlled and exerted during the SCN CC-controlled active phase, have been associated with increased risk of developing diabetes (hypoinsulinemia, hyperglycemia, reduced glucose tolerance), high free fatty acid (FFA) level, hypertriglyceridemia, obesity, and metabolic syndrome (6–9, 21–23). It is, however, largely unknown how these pathologies are mechanistically related to the PCC/SCN CC misalignments (3, 6–9). Most interestingly, our present study reveals, at the molecular level, that all of these metabolic pathologies eventually develop in RF mice, as a consequence of the RF-induced 12-h misalignment between the PCCs and rest and active phases of activity of the SCN CC.

A Misalignment-Induced Decrease in INS Production Leads to Diabetes in RF Mice.

Under ad libitum feeding during the active phase (the ZT8–ZT12 period; Fig. 4A), the Bmal1 activity of the pancreatic CC transcribes INS, insulin receptor (INSR), insulin receptor substrate 2 (IRS2), and glucose transporter 2 (GLUT2) genes, thus aligning INS synthesis and signaling with the ZT12 feeding time (Fig. 3I) (24) and enabling maximal postprandial (ZT12–ZT16) glucose-stimulated INS secretion (GSIS) (Fig. 3A). This GSIS (step 1 in Fig. 4A), at the beginning of the active phase, is critical to prevent lipolysis within the adipose tissue (11), thereby decreasing the FFA blood level (Fig. 3A and steps 9–10 in Fig. 4A). Moreover, the INS blood increase (i) counteracts glucagon secretion to prevent pCREB activity in liver (steps 2–4 in Fig. 4A) and (ii) activates pAKT (Fig. 1E and Fig. S1L) to inactivate the FOXO1 protein in liver (11) (steps 5–7 in Fig. 4A). This combined decrease in pCREB and FOXO1 reduces the PEPCK expression in liver, thereby preventing hepatic gluconeogenesis (ZT16–ZT4) and maintaining the physiological glucose blood level (step 8 in Fig. 4A), which ensures glucose tolerance (steps 11–13 in Fig. 4A).

Open in a separate window

Fig. 3.

Prolonged feeding during the rest phase leads to the development of diabetes and fatty liver. (A) Levels of blood components, as indicated, after RF15, RF30, and RF90. (B) As in A, but measuring TG levels in liver. (C) As in A, but measuring blood levels of total, HDL, and LDL cholesterol. (D) As in A, but measuring total bile acids. (E) Oil red O staining to detect TG deposition in control and RF90 liver. (F–H) Glucose tolerance tests (GTTs) and GSIS after RF15 (F), RF30 (G), and RF90 (H). (I) RNA transcript levels of genes, as indicated, in pancreas of control and RF30 mice. (J) RNA transcript levels of genes as indicated in liver of control and RF30 mice. (K) Total body weight and weight of the peritoneal adipose tissue in control and RF mice. (L) Insulin tolerance tests (ITT) after RF15, RF30, and RF90 days. All values are mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001.

Open in a separate window

Fig. 4.

A misalignment of PCCs with the SCN CC under RF regime progressively induces a metabolic syndrome. (A) Schematic representation of how PCCs alignment with the SCN CC maintains INS, glucose, and FFA homeostasis. (B) A schematic representation of how PCCs misalignment with the SCN CC in RF mice leads to hypoinsulinemia, hyperglycemia, and increased FFA level. (C) A schematic representation of how PCCs alignment with the SCN CC maintains lipogenesis and TG level and prevents adiposity. (D) A schematic representation of how PCCs misalignment with the SCN CC in RF mice leads to an increase in lipogenesis, hypertriglyceridemia, and adiposity.

We found by RF7, that the CC shift was complete in pancreas, which resulted in a permanent misalignment in the expression of genes critically involved in insulin synthesis and signaling (see below) and led to a recurrent hypoinsulinemia (in RF mice) during the active phase, thereby accounting for all of the pathological metabolic features typically associated with diabetes (Fig. 4B). Due to this RF CC shift, Bmal1-dependent transcriptions in pancreas were shifted to the ZT16–ZT20 period (Fig. 3I), at a time where dietary glucose is unavailable due to the rest phase feeding at ZT0 (Fig. 4B).The ensuing RF-hypoinsulinemia (ZT12–ZT0; Fig. 3A) induced the FOXO1 activity in pancreas, as a consequence of a reduction in pAKT level (11, 25, 26), thus leading to a repression (25) of the transcription of the INS gene at ZT0 (steps 14–16 in Fig. 4B and Fig. 3I) and to a decreased postprandial GSIS in RF mice (Fig. 3A). Moreover, as a result of the RF hypoinsulinemia, there was an increase in PEPCK expression due to enhanced pCREB and FOXO1 activity (Fig. S3D and steps 2–7 in Fig. 4B), which resulted in RF hyperglycemia (Fig. 3A and step 8 in Fig. 4B). Taken together, this RF hypoinsulinemia and hyperglycemia account for the reduced glucose tolerance in RF mice (Fig. 3 F–H and steps 11–13 in Fig. 4B). However, even at RF90, an INS-resistant state was not achieved, although there was a progressive deterioration from RF15 to RF90 (Fig. 3L). Importantly, the reduction in INS level in RF mice also accounted for the inability of these mice to suppress lipolysis in adipose tissue and the resulting increase in FFA blood level (Fig. 3A and steps 9–10 in Fig. 4B).

Open in a separate window

Fig. S3.

Metabolic features of RF30 mice. (A) RNA transcript levels of genes, as indicated, in liver of control and RF30 mice. (B) RNA transcript levels of CC components in IECs of control and RF30 mice. (C) RNA transcript levels of CC components liver of control and RF30 mice. (D) RNA transcript levels of genes, as indicated, in liver of control and RF30 mice. (E) RNA transcript levels of HSD3B5 and IGFBP1 in liver of control and RF30 mice. (F) FGF21 and IGF1 levels in blood of control and RF30 mice. All values are mean ± SEM. *P < 0.05.

In summary, taken together, our present results establish how an RF-induced misalignment of PCCs and SCN CC triggers the development of diabetes (hypoinsulinemia, hyperglycemia, reduced glucose tolerance) and increased FFA level through misexpression during the active phase of PCC genes normally expressed during the rest phase.

Isolation and Laser Capture Microdissection of SCN and PVN.

The brains were isolated and frozen in iso-pentane (Sigma Aldrich) kept on dry ice and immediately stored at −70 °C. The frozen brains were fixed by Cryomatrix (Thermo Scientific), and 12-µm-thick sections were cut using a cryotome. The sections were mounted on PET-Membrane frame slides (Leica no. 11505151), which were first fixed in 70% (vol/vol) ethanol for 1 min and then stained in cresyl violet (Sigma Aldrich) for 3 min, followed by rinsing in 70% and 100% ethanol for 30 s each. Finally, the slides were air dried and used for laser capture microdissection. The locations of the SCN and PVN was confirmed using the brain atlas. The laser-cut SCN and PVN samples were collected in the caps of RNase-free 200-µL PCR tubes containing RLT buffer (a guanidine-thiocyanate–containing lysis buffer) (Qiagen). RNA was extracted using the RNeasy micro kit (Qiagen) following the manufacturer’s instruction manual. A standard reverse transcription protocol was used to convert the isolated RNA into cDNA, and purity of SCN samples was verified by measuring the transcripts for arginine vasopressin (AVP) and vasoactive intestinal polypeptide (VIP), expressed relative to GAPDH.

In Situ Hybridization.

In situ hybridization was performed using digioxigenin (DIG)-labeled riboprobes as previously described (35). PPARα and RORα cDNA was PCR amplified from mouse liver and subsequently cloned into the pGEM-T easy vector system (Promega) and sequenced to verify the gene identify. DIG-labeled sense and antisense riboprobes were synthesized using the in vitro transcription kit (Promega). Brain sections (14 µm thick) were cut and mounted on superfrost plus (Thermo Scientific) slides. The sections were stored at −80 °C. For in situ hybridization, the slides were first defrosted and then air dried at room temperature for 30 min. The sections were postfixed in ethanol gradation of 70%, 90%, and 100% for 5 min each and finally rinsed in PBS. The sections were hybridized overnight using 0.1 µg/mL RORα riboprobe and 1 μg/mL PPARα riboprobe in hybridization solution and incubated at 65 °C in a humidified chamber. The slides were washed in 5× SSC for 15 min followed by 0.2× SSC for 2 h at 72 °C. The blocking was performed for 2 h in blocking solution [maleic acid buffer (MAB), pH 7.5, Tween-20 0.1%, blocking reagent 2% (Roche), and 20% heat inactivated normal goat serum]. Next, alkaline phosphatase-conjugated anti-DIG Fab fragment (Roche) was added on the glass sides at a dilution of 1:2,000 in blocking solution and incubated overnight at 4 °C under glass coverslip. The slides were then washed five times in MAB with 10% Tween-20 for 5 min each and finally rinsed in alkaline phosphatase buffer. Nitro blue tetrazolium (NBT)/5-Bromo 4-chloro-3-indolyl phosphate (BCIP) (Roche) was used as staining solution for developing slides to detect alkaline phosphatase activity. The sections were incubated in dark for color development for 1 d (RORα) and 3 d (PPARα) at room temperature. After color development, the slides were rinsed in PBS, air dried, and mounted in PERTEX. Photographs were taken using Leica M420 macroscope and DMLB/DM4000B microscopes equipped with Photometrics digital cameras and CoolSnap imaging software (Roger Scientific). Primer sequences for the preparation of riboprobes are available on request.

RNA Transcript Determination.

Freshly isolated IECs, liver, pancreas, and hippocampus samples were used for RNA isolation using TRI reagent (Molecular Research Center). Subsequent to the verification of RNA quality (gel electrophoresis), 1 μg total RNA (determined spectrophotometrically) was reverse transcribed using random hexamers and SuperScript II reagents (Invitrogen) per the manufacturer’s instructions. The synthesized cDNA was used for qRT-PCR with SYBRgreen (QIAGEN) and expressed relative to the hypoxanthine phosphoribosyltransferase (HPRT) levels, as previously described (19). Primer sequences are available on request. For transcript determination, four control and four RF mice were killed at each time point, and each experiment was replicated three times.

Protein Immunoblots.

Immunoblots from isolated liver were performed following standard SDS/PAGE procedures. Proteins were visualized following ECL (Pierce), and images were captured using CCD. The primary antibodies for IRS1 (2390), pIRS1 Ser307 (2381), pAkt Ser473 (4060), Akt (4685), pGSK3β Ser9 (5558), GSK3β (12456), and pRevErbα Ser55/59 (2129) were obtained from Cell Signaling Technology. RevErbα (ab115552) and Bmal1 (ab93806) antibodies were obtained from abcam. GAPDH antibody was from Millipore.

FGF21 Measurement.

FGF21 was assayed from the plasma samples obtained from the blood of control and RF mice (eight animals were used per group), using the FGF21 quantikine ELISA kit (R&D; MF2100) per the manufacturer’s instructions.

ChIP Assays.

ChIP was performed as reported (19) with some minor modifications. Briefly, isolated IEC suspension in PBS was cross-linked with 1% formaldehyde for 15 min at room temperature; cross-linking was stopped by addition of 2 M glycine (0.125 M final concentration) at room temperature for 5 min. Cells were pelleted, and 500 μL lysis buffer was added in presence of protease inhibitors (Roche) on ice. For liver samples, identical lobes of liver from different groups of mice were disrupted using a dounce homogenizer; samples were then cross-linked and processed as described above. For adrenal glands, 12 adrenal glands were pulled and homogenized for each antibody per time point. Following cell lysis, the samples were sonicated (Bioruptor; Diagenode) to generate fragments of average length of 200–500 bp. Cellular debris were removed by centrifugation at 4 °C for 10 min (10,000 × g), and supernatant was precleared with Protein A/G-Sepharose (Roche) beads and preblocked with salmon DNA and BSA for 60 min at 4 °C. Beads were pelleted and discarded; 10% of the lysate was stored from each sample as the source of Input, and the remaining lysate was diluted eight times (five times for adrenal glands) in dilution buffer [16.7 mM TrisHCl, pH 8.1, 0.01% (wt/vol) SDS, 1.1% (vol/vol) Triton X-100, 1.2 mM EDTA, 16.7 mM NaCl, protease inhibitor mixture] in the presence of different primary antibodies for 14 h at 4 °C, on a flip-flop rocker. Ninety microliters Protein A/G-Sepharose beads (preblocked with salmon DNA and BSA) was then added for 90 min at 4 °C. Immunecomplexes were recovered by centrifugation at 500 × g for 1 min. Beads were washed extensively at 4 °C in low salt buffer [20 mM TrisHCl, pH 8.1, 0.1% (wt/vol) SDS, 1% (vol/vol) Triton X-100, 2 mM EDTA, 150 mM NaCl], in high salt buffer [20 mM TrisHCl, pH 8.1, 0.1% (wt/vol) SDS, 1% (vol/vol) Triton X-100, 2 mM EDTA, 500 mM NaCl], in LiCl buffer [10 mM TrisHCl, pH 8.1, 250 mM LiCl, 1% (vol/vol) Nonidet P-40, 1% (wt/vol) sodium deoxycholate, 1 mM EDTA], and finally in 1 mL TE buffer (10 mM TrisHCl, pH 8.0, 1 mM EDTA). The bound chromatin was released from the beads by intermittent vortexing at room temperature in 200 μL elution buffer [1 (wt/vol) SDS and 100 mM NaHCO₃]. One milliliter of 10 mg/mL RnaseA and 5 M NaCl (200 mM final concentration) was added to the eluate and incubated overnight at 65 °C and then treated with Proteinase K for 1 h at 55 °C; DNA was purified using the QIAGEN PCR purification kit in a final volume of 50 μL. The qPCR was done using this eluted DNA and SYBRgreen reagent (QIAGEN). PCR cycles were verified to be within the linear range of amplification. Primer sequences are available on request.

Plasma Metabolic Analysis.

Blood glucose, INS, TG, FFA, cholesterol (total, LDL, HDL), bile acids, and IGF1 levels were measured from control and RF mice (eight animals per group per time point) at indicated ZTs. Blood glucose levels were determined on blood collected from the tail vein using a handheld Accu-check active glucometer (Roche). For all other measurements, blood were collected by retro-orbital puncture in EDTA-coated vials, plasma was separated, and measurements were done in the metabolomics unit of the IGBMC/ICS.

GTTs and ITTs.

GTTs and ITTs were performed after 15, 30, and 90 d of RF. Blood glucose was measured at the indicated times after i.p. administration of glucose (1 g/kg) of body weight. GSIS was also similarly studied after i.p. administration of glucose. ITTs were carried out after i.p. administration of 0.75 U INS/kg of body weight (Humulin R; EliLilly). INS level at the beginning of the experiment was considered as 100%. Eight to 10 control and RF mice were used for GTT and ITT experiments. GTT and GSIS were performed from the same animals. For glucose administration-dependent reversal of RF PCC shift, glucose (2 g/kg) was i.p. administered for 4 consecutive d at ZT18.

Pharmacological Inhibition of GSK3β.

GSK3β activity was inhibited by i.p. administration of either LiCl (Merck) or AR-A014418 (ARA; Sigma Aldrich; 3230) at ZT18 to RF mice. LiCl (25 mEq/kg) was prepared in PBS, whereas ARA (8 mg/kg) was dissolved in DMSO (Sigma Aldrich) and diluted in PBS before injection. Both inhibitors were administered for 5 consecutive RF days (injected every day at ZT18). One group of RF mice were mock i.p. administered with DMSO + PBS.

Oil Red O Staining.

Following death, liver samples were frozen in Cryomatrix and kept at −20 °C. Eight-micrometer sections were prepared and stained in 0.5% Oil red O stain (prepared in glycerol) for lipid (TG) and then in hematoxylin for 5 s, and images were captured.

Quantification of TG Levels in Liver.

Hepatic TG levels were analyzed as per a previous report (33). Briefly liver extracts were prepared by homogenization in 0.25% sucrose with 1 mmol/L EDTA, and lipids were extracted using chloroform/methanol [2:1 (vol/vol)] and suspended with 5% fatty acid-free BSA. TG level was measured using triglyceride assay reagents (Sigma). The assay was performed in the metabolomics unit of ICS.

Circadian Locomotor Activity and Ingestive Behaviors.

Spontaneous locomotor activity of control and RF mice in the front and rear portions of the cages were simultaneously recorded from 24 individual boxes equipped with infrared captors, connected to computers. The quantity of water and food consumed was measured during the test period using an automated pellet feeder and lickometer (Imetronic, Pessac, France). For RF cages, the automated pellet feeder was disconnected manually at 6:00 PM (ZT12), whereas it was connected back to the cages at 6:00 AM (ZT0). Mice were tested for 32 h to measure habituation to the apparatus, as well as nocturnal and diurnal activities, following which the locomotor activity data are recorded for 10 consecutive light-dark periods. Results are expressed per 1-h periods.

We thank the staffs of animal house facilities in Institut de Génétique et de Biologie Moléculaire et Cellulaire (IGBMC)/Institut Clinique de la souris (ICS) for help. This work was supported by the Centre national de la recherche scientifique (CNRS), Institut national de la santé et de la recherche médicale (INSERM), University of Strasbourg Institute for Advanced Studies, and the Association pour la Recherche a l’IGBMC (ARI). A.M., A.K., M.D., and N.M. were supported by fellowships from ARI.